@inproceedings{boulmerka_thresholding-based_2012,

title = {Thresholding-based segmentation revisited using mixtures of generalized Gaussian distributions},

author = {A. Boulmerka and M. S. Allili},

url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84874575463&partnerID=40&md5=0665cce9aa19af524d1213c1ff728d94},

isbn = {978-4-9906441-0-9},

year  = {2012},

date = {2012-01-01},

booktitle = {Proceedings - International Conference on Pattern Recognition},

pages = {2894–2897},

address = {Tsukuba},

abstract = {This paper presents a new approach to image-thresholding-based segmentation. It considerably improves existing methods by efficiently modeling non-Gaussian and multi-modal class-conditional distributions. The proposed approach seamlessly: 1) extends the Otsu's method to arbitrary numbers of thresholds and 2) extends the Kittler and Illingworth minimum error thresholding to non-Gaussian and multi-modal class-conditional data. We use the recently-proposed mixture of generalized Gaussian distributions (MoGG) modeling, which enables to efficiently represent heavy-tailed data, as well as multi-modal histograms with flat and sharply-shaped peaks. Experiments performed on synthetic data and real-world image segmentation show the performance of the proposed approach with comparison to recent state-of-the-art techniques. © 2012 ICPR Org Committee.},

note = {ISSN: 10514651},

keywords = {Arbitrary number, Gaussian noise (electronic), Generalized Gaussian Distributions, Heavy-tailed, Image segmentation, Kittler, Minimum error thresholding, Multi-modal, New approaches, Non-Gaussian, Otsu's method, Pattern Recognition, State-of-the-art techniques, Synthetic data},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{allili_unsupervised_2010,

title = {Unsupervised feature selection and learning for image segmentation},

author = {M. S. Allili and D. Ziou and N. Bouguila and S. Boutemedjet},

url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-77954407977&doi=10.1109%2fCRV.2010.44&partnerID=40&md5=a7d8e3147216429f18ef7af3167acb42},

doi = {10.1109/CRV.2010.44},

isbn = {978-0-7695-4040-5},

year  = {2010},

date = {2010-01-01},

booktitle = {CRV 2010 - 7th Canadian Conference on Computer and Robot Vision},

pages = {285–292},

address = {Ottawa, ON},

abstract = {In this paper we investigate the integration of feature selection in segmentation through an unsupervised learning approach. We propose a clustering algorithm that efficiently mitigates image under/over-segmentation, by combining generalized Gaussian mixture modeling and feature selection. The algorithm is based on generalized Gaussian mixture modeling which is less prone to region number over-estimation in case of noisy and heavy-tailed image distributions. On the other hand, our feature selection mechanism allows to automatically discard uninformative features, which leads to better discrimination and localization of regions in high-dimensional spaces. Experimental results on a large database of real-world images showed us the effectiveness of the proposed approach. © 2010 IEEE.},

keywords = {Clustering algorithms, Computer vision, Evolutionary algorithms, Feature extraction, Feature selection, Gaussian distribution, Generalized Gaussian, Generalized Gaussian Distributions, Heavy-tailed, High dimensional spaces, Image distributions, Image segmentation, Large database, Over-estimation, Real-world image, Unsupervised feature selection},

pubstate = {published},

tppubtype = {inproceedings}

}